List your company

Time to First Token (TTFT) on NVFP4-W4A16

182Time to First Token (ms)FlashRT
-640.644,912.1810,46516,017.82Jun 18, 2026

Evaluation results

MethodLinks
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=4096, Reuse Strategy=with reuse (capsule)2026.06182
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=2048, Reuse Strategy=with reuse (capsule)2026.06183
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=8192, Reuse Strategy=with reuse (capsule)2026.06192
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=16384, Reuse Strategy=with reuse (capsule)2026.06202
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=4096, Reuse condition=with reuse (capsule)2026.06225
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=16384, Reuse condition=with reuse (capsule)2026.06236
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=2048, Reuse condition=with reuse (capsule)2026.06240
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=8192, Reuse condition=with reuse (capsule)2026.06244
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=4096, Reuse Strategy=with reuse (APC)2026.06302
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=4096, Reuse condition=with reuse (APC)2026.06332
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=8192, Reuse Strategy=with reuse (APC)2026.06482
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=2048, Reuse Strategy=with reuse (APC)2026.06571
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=8192, Reuse condition=with reuse (APC)2026.06586
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=2048, Reuse condition=with reuse (APC)2026.06663
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=16384, Reuse Strategy=with reuse (APC)2026.06839
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=2048, Reuse Strategy=no reuse (cold)2026.06906
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=16384, Reuse condition=with reuse (APC)2026.061,036
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=4096, Reuse Strategy=no reuse (cold)2026.061,652
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=2048, Reuse Strategy=no reuse (cold)2026.062,122
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=2048, Reuse condition=no reuse (cold)2026.062,196
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=2048, Reuse condition=no reuse (cold)2026.062,529
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=8192, Reuse Strategy=no reuse (cold)2026.063,221
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=4096, Reuse Strategy=no reuse (cold)2026.064,056
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=4096, Reuse condition=no reuse (cold)2026.064,208
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=4096, Reuse condition=no reuse (cold)2026.065,197
FlashRTHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=16384, Reuse Strategy=no reuse (cold)2026.066,613
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=8192, Reuse Strategy=no reuse (cold)2026.067,935
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=8192, Reuse condition=no reuse (cold)2026.068,502
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=8192, Reuse condition=no reuse (cold)2026.0610,121
vLLMHardware=NVIDIA DGX Spark (GB10, sm_121), Prefix=16384, Reuse Strategy=no reuse (cold)2026.0616,129
FlashRTHardware=Jetson AGX Thor (SM110), Prefix length=16384, Reuse condition=no reuse (cold)2026.0617,922
vLLMHardware=Jetson AGX Thor (SM110), Prefix length=16384, Reuse condition=no reuse (cold)2026.0620,748